Automatic Machine Translation Evaluation in Many Languages via Zero-Shot Paraphrasing

@inproceedings{Thompson2020AutomaticMT,
  title={Automatic Machine Translation Evaluation in Many Languages via Zero-Shot Paraphrasing},
  author={Brian Thompson and Matt Post},
  booktitle={EMNLP},
  year={2020}
}
We propose the use of a sequence-to-sequence paraphraser for automatic machine translation evaluation. The paraphraser takes a human reference as input and then force-decodes and scores an MT system output. We propose training the aforementioned paraphraser as a multilingual NMT system, treating paraphrasing as a zero-shot "language pair" (e.g., Russian to Russian). We denote our paraphraser "unbiased" because the mode of our model's output probability is centered around a copy of the input… Expand
10 Citations
Learning to Evaluate Translation Beyond English: BLEURT Submissions to the WMT Metrics 2020 Shared Task
  • 1
  • PDF
Results of the WMT20 Metrics Shared Task
  • 4
  • Highly Influenced
  • PDF

References

SHOWING 1-10 OF 98 REFERENCES
Machine Translation Evaluation using Bi-directional Entailment
  • 3
  • PDF
Learning Paraphrastic Sentence Embeddings from Back-Translated Bitext
  • 51
  • PDF
Improving Zero-shot Translation with Language-Independent Constraints
  • 11
  • PDF
Massively Multilingual Sentence Embeddings for Zero-Shot Cross-Lingual Transfer and Beyond
  • 287
  • PDF
Universal Neural Machine Translation for Extremely Low Resource Languages
  • 140
  • PDF
Analyzing Uncertainty in Neural Machine Translation
  • 97
  • PDF
Language Models are Unsupervised Multitask Learners
  • 2,628
  • PDF
...
1
2
3
4
5
...